Method for sensor-free determination of the rotor position of electronically commutated multiple-phase synchronous machines

Abstract

The invention relates to a method for sensor-free position determination of the rotor position of an electronically commutated multiple-phase EC motor comprising a rotor, and a stator and comprising a commutation device for generating string currents in the coil system of the stator by applying a test signal in the coil system, measuring the current value i in the strand during the measurement phase (PHMess) as a current response to the test signal, calculating i for determining the envelope determining of the current response and determining the rotor position therefrom.

The invention claimed is: 
     
       1. A method for sensor-free determination of a rotor position of an electronically commutated multiple-phase EC motor with a rotor, a stator and including a commutation device for generating phase currents in a coil system of the stator, the method comprising:
 a. applying a test signal in the coil system, wherein an amplitude A of the test signal has an envelope curve variation which at first rises during a rising phase (PH an ) and then remains constant for a definite number of periods during a measurement phase (PH Mess ), 
 b. measuring a current value i in a strand during the measurement phase (PH Mess ) as a current response to the test signal, 
 c. calculating the envelope curve of the current response and 
 d. determining the rotor position (γ 0 ) of a rotor with a rotor diameter r Rotor  of a fan driven by the EC motor with an impeller diameter D Ventilator , a minimum number (N A ) of periods is determined as a rounded-off quotient of the impeller half-diameter D Ventilator  and the rotor diameter r Rotor , by evaluating a curve maxima of the envelope curve of the current response. 
 
     
     
       2. The method as claimed in  claim 1 , wherein the test signal is a test voltage signal (u 1 ) or a test current signal (i 1 ). 
     
     
       3. The method as claimed in  claim 1 , wherein, during step a), applying of the test signal u 1  is done in the coil system, wherein the test signal u 1  constitutes a voltage vector with a radian frequency (ω) and the voltage amplitude A of the voltage vector at first rises during a rising phase (PH an ) to a test voltage and then remains constant for a definite number of periods (N A ) during a measurement phase (PH Mess ). 
     
     
       4. The method as claimed in  claim 3 , wherein the amplitude of the voltage vector moreover follows the time-dependent function (f D ) whose value range lies between 0 and 1:
 f D =0 before the rising phase (PH an ) 
 f D =f(t) with f′(t)>0 during the rising phase (PH an ) 
 f D =1 during the measurement phase (PH Mess ). 
 
     
     
       5. The method as claimed in  claim 3 , wherein the amplitude of the voltage vector moreover follows the time-dependent function (f D ):
 f D =f(t) with f′(t)<0 during the falling phase (PH ab ). 
 
     
     
       6. The method as claimed in  claim 5 , wherein, in order to ensure a sufficiently long rise time to avoid disruptive noises for the rising phase (PH an ), a minimum number (N A ) of periods is selected, preferably at least 2 periods. 
     
     
       7. The method as claimed in  claim 1 , wherein the rise in the test signal during the rising phase (PH an ) has a variation curve such that it is at least once continuously differentiable in the entire variation. 
     
     
       8. The method as claimed in  claim 7 , wherein a function f(t) is a linear function, a sin 2  function, or a sigmoid function. 
     
     
       9. The method as claimed in  claim 1 , wherein, at an end of the measurement phase (PH Mess ), there occurs either a falling phase (PH ab ), during which the amplitude A of the envelope curve of the test signal is further reduced according to the variation of the rise function. 
     
     
       10. The method as claimed in  claim 1 , wherein, at an end of the measurement phase (PH Mess ), there occurs a direct transition to the operation of the EC motor by commutation of the phase currents of the rotor, after determination of the rotor position in step d). 
     
     
       11. The method as claimed in  claim 1 , wherein the amplitude rise A of the test voltage signal is linear or rises according to a sin 2  function or a sigmoid function (S-function). 
     
     
       12. The method as claimed in  claim 1 , wherein the determination of the rotor position (γ 0 ) in step d) is done by polling a relative angle value (θ) of a voltage vector at which an amplitude value of a current vector (i) of the current response has a maximum or by direct calculation in a stator-fixed reference system by determining an arc-tangent of a quotient of directional components of the current vector (i). 
     
     
       13. The method as claimed in  claim 1 , wherein an internal voltage compensation is provided for determining the rotor position (γ 0 ) in step d), wherein for each phase current one determines a mean value of the phase current from every two current maxima and at the end of the period subtracts it from a memorized phase current variation. 
     
     
       14. A method for sensor-free determination of a rotor position of an electronically commutated multiple-phase EC motor with a rotor, a stator and including a commutation device for generating phase currents in a coil system of the stator, the method comprising:
 a. applying a test signal in the coil system, wherein an amplitude A of the test signal has an envelope curve variation which at first rises during a rising phase (PH an ) and then remains constant for a definite number of periods during a measurement phase (PH Mess ), 
 b. measuring a current value i in a strand during the measurement phase (PH Mess ) as a current response to the test signal, 
 c. calculating the envelope curve of the current response and 
 d. determining the rotor position γ 0  of a rotor with a rotor diameter r Rotor  of a fan driven by the EC motor with an impeller diameter D Ventilator , a minimum number N A  of periods is determined as a rounded-off quotient of the impeller half-diameter D Ventilator  and the rotor diameter r Rotor , by evaluating a curve maxima of the envelope curve of the current response; and 
 e. determining a radian frequency ω=2πf TEST  of the voltage vector according to an excitation frequency between a minimum and maximum frequency, which is determined by means of frequency analysis.